Abstract: This application relates to the field of intelligent vehicle technologies in the field of artificial intelligence technologies, and in particular, to a controller system. The controller system includes: an intelligent driving domain control unit; a human-machine interaction domain control unit; and a sensor interface unit that is connected to the intelligent driving domain control unit and the human-machine interaction domain control unit. The sensor interface unit is connected to a sensor, and transmit data of the sensor to the intelligent driving domain control unit and the human-machine interaction domain control unit. In this application, external cable connections can be simplified, occupation of data stream transfer resources can be reduced, and power consumption of a vehicle can be reduced.

Abstract: An in-vehicle computing apparatus comprises a mainboard, a heat dissipation component, and a target printed circuit board (PCB). A system on chip (SOC) and a connection component are disposed on the target PCB. A side of the mainboard is attached to the heat dissipation component. The target PCB is located on a side of the heat dissipation component that faces away from the mainboard, and is detachably connected to the heat dissipation component by using the connection component.

Abstract: An electronic rearview mirror system includes a plurality of cameras, a plurality of displays, a first controller, and a second controller. The plurality of cameras is configured to collect an image at a left rearview angle and an image at a right rearview angle outside a vehicle. The first controller and the second controller are configured to: obtain information about the images from the plurality of cameras and perform image processing on the information. The first controller and the second controller are further configured to serve as controllers of an in-vehicle system different from the electronic rearview mirror system. The plurality of displays is configured to display images obtained after the image processing.

Abstract: An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Abstract: Example positioning methods and apparatus are described. One example positioning system includes a first global navigation satellite system (GNSS) antenna, an in-vehicle communication apparatus, and a controller. The first GNSS antenna is configured to obtain satellite data, and transmit the satellite data to the in-vehicle communication apparatus. The in-vehicle communication apparatus is configured to determine first positioning information based on at least one of the satellite data or first inertia measurement unit (IMU) data, and transmit the satellite data to the controller. The controller is configured to determine second positioning information based on the satellite data and second IMU data. Precision of the second positioning information is higher than precision of the first positioning information.

Abstract: An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Abstract: Provided is a black box data management method and apparatus for an intelligent driving vehicle. A black box device first obtains black box data based on a black box trigger event, wherein the black box device is configured to manage the black box data in the intelligent driving vehicle. The black box device evaluates a storage level of the black box data based on an event type of the black box trigger event and a data type of the black box data. The black box device stores the black box data based on the storage level and according to a preset rule.

Abstract: A time synchronization method and apparatus includes determining a time difference between reference time and system time of an artificial intelligence device, where the reference time is timed by an internal clock of the artificial intelligence device and is aligned based on a satellite timing signal, or the reference time is timed by an internal clock of the artificial intelligence device; and adjusting the system time based on a preset step value if the time difference is greater than a preset value.

Abstract: This application relates to the field of intelligent vehicle technologies in the field of artificial intelligence technologies, and in particular, to a controller system. The controller system includes: an intelligent driving domain control unit; a human-machine interaction domain control unit; and a sensor interface unit that is connected to the intelligent driving domain control unit and the human-machine interaction domain control unit. The sensor interface unit is connected to a sensor, and transmit data of the sensor to the intelligent driving domain control unit and the human-machine interaction domain control unit. In this application, external cable connections can be simplified, occupation of data stream transfer resources can be reduced, and power consumption of a vehicle can be reduced.

Abstract: An in-vehicle computing apparatus comprises a mainboard, a heat dissipation component, and a target printed circuit board (PCB). A system on chip (SOC) and a connection component are disposed on the target PCB. A side of the mainboard is attached to the heat dissipation component. The target PCB is located on a side of the heat dissipation component that faces away from the mainboard, and is detachably connected to the heat dissipation component by using the connection component.

Abstract: An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Abstract: A time synchronization method and apparatus includes determining a time difference between reference time and system time of an artificial intelligence device, where the reference time is timed by an internal clock of the artificial intelligence device and is aligned based on a satellite timing signal, or the reference time is timed by an internal clock of the artificial intelligence device; and adjusting the system time based on a preset step value if the time difference is greater than a preset value.

Abstract: This application discloses a black box data management method for an intelligent driving vehicle. A black box device first obtains black box data based on a black box trigger event, evaluates a storage level of the black box data based on an event type of the black box trigger event and a data type; and stores the black box data based on the storage level and according to a preset rule. A method is disclosed for recording and storing the black box data in the intelligent driving vehicle, thereby improving the effectiveness of black box data of the intelligent driving vehicle and enhancing vehicle safety.

Abstract: An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Abstract: A vehicle control method comprising: obtaining first vehicle data and second vehicle data, processing the first vehicle data using the first computing system to obtain first structured data; processing the second vehicle data using the second computing system to obtain second structured data; and further controlling safe driving of a vehicle based on the first structured data and the second structured data, wherein the first structured data is used to represent an environment that is of the vehicle at a first time point and that is detected by the first group sensing apparatus, and the second structured data is used to represent an environment that is of the vehicle at the first time point and that is detected by the second group sensing apparatus.

Abstract: A vehicle control method comprising: obtaining first vehicle data and second vehicle data, processing the first vehicle data using the first computing system to obtain first structured data; processing the second vehicle data using the second computing system to obtain second structured data; and further controlling safe driving of a vehicle based on the first structured data and the second structured data, wherein the first structured data is used to represent an environment that is of the vehicle at a first time point and that is detected by the first group sensing apparatus, and the second structured data is used to represent an environment that is of the vehicle at the first time point and that is detected by the second group sensing apparatus.

Abstract: An acceleration resource scheduling method includes: receiving an acceleration instruction sent by the virtual machine, where the acceleration instruction includes to-be-accelerated data; determining a virtual accelerator allocated to the virtual machine; determining, based on the virtual accelerator, a network accelerator that is to process the acceleration instruction, and sending the acceleration instruction to the network accelerator, so that the network accelerator sends the acceleration instruction to a physical accelerator that is to process the acceleration instruction; receiving a computing result that is returned after the physical accelerator performs acceleration computing on the to-be-accelerated data by using the physical acceleration resource; and sending the computing result to the virtual machine.

Abstract: A computer system and a storage access apparatus are provided. The storage access apparatus is configured to: configure at least one virtual function VF by using a physical function PF of an single-root I/O virtualization (SR-IOV) function, and configure a VM-VF association relationship, so that a direct access path is established between a VM and a VF that are associated; and obtain, by using a network interface, data block resources provided by at least one network storage device that is connected to the storage access apparatus, form a plurality of virtual volumes by using the obtained data block resources, and configure a VF-virtual volume association relationship. Therefore, by using a storage access method supported by the storage access apparatus, a storage access path and latency can be shortened, and occupation of CPU resources in a compute node can be reduced.